138

INFRA-RED TRANSMISSION AND REFRACTION DATA

to the yellow helium line as a basis. After this the bolometer or thermopile is adjusted upon the yellow helium line, then on rotating the spectrometer through a certain angle, say 2°, the corresponding wave-length is, say, 6,u, while a rotation of 4° places the bolometer at about 8-1/j. in the spectrum of a 60° fluorite prism ((6), p. 49).

Elimination of Scattered Radiation in Spectral Energy Measurements.—In the design of optical instruments there are opportunities for great improvement in this respect. Take, for example, the image-forming telescope of a spectrometer. The telescope tube should be large and suitably diaphragmed so that when the violet end of the spectrum is incident upon the radiometer receiver the infra - red end of the spectrum cannot be reflected from the side of the tube and impinge upon the receiver. Furthermore, the bevelled edges of the exit slits of the spectrometer should face outwards (18) instead of facing the image-forming lens, as obtains in commercial instruments.

By using suitably constructed optical instruments the scattered radiation is practically eliminated. What .little remains may be obviated by using, before the entrance slit of the spectrometer, a shutter (22, 5, 18), which is opaque to the region of the spectrum under investigation, but which transmits the scattered radiations. In this manner the scattered radiations are incident upon the radiometer all the time and, hence, do not affect the energy measurements. Using a spectrometer which ia provided with slits and diaphragms, as just mentioned, it has been found (4) that the scattered radiation was immeasurable in comparison with the intensities under investigation.

§ (4) OPTICAL CONSTANTS OF GLASS.—In the ultra - violet end of the spectrum ordinary crown glass is transparent to about 0-3^, while the flint-silicate glasses absorb strongly throughout the blue and violet end of the spectrum.

In the infra-red spectrum all glasses (38, 40) begin to absorb at about 2/z, and for a thickness of 1 cm. they are practically opaque to radiations of wave-length greater than 3/*. Glasses containing traces of iron impurities have an absorption band at 1/j,.

The refractive indices of various glasses have been determined by Rubens (20). He determined the refractive indices also of water, xylol, benzol, etc. However, in view of the fact that the refractive indices depend upon the composition of the glass, no refraction data are given in this paper. Practically no infra-red work is being done with glass prisms.

§ (6) OPTICAL CONSTANTS OF CARBON BISULPHIDE.—Carbon disulphide is quite transparent in the infra-red. In the region to 3,u Rubens (20)

found an absorption of only 5 to 7 per cent for a 1 cm. thickness. Beyond 4/j. there are a number of very large absorption bands (36).

As illustrated in Fig. 3, carbon disulphide has a very much larger dispersion than quartz, etc., in the region of 0-5 to 2/j., and hence is especially adapted for certain fields of spectro-radiometry.

The infra-red refractive indices of carbon disulphide were determined by Rubens (20). Those in the visible and in the ultra-violet were determined by Flatow (32). Rubens's values of the infra-red refractive indices are given in Table I.

TABLE I INDICES OF REFRACTION OF CARBON BISULPHIDE

IN AlR AT 15° C. (R0BEN8)

§ (6) OPTICAL CONSTANTS OF QUARTZ.— Quartz is one of the most useful materials for prisms. It is extremely transparent to ultraviolet radiations. Pfliiger (41) found a transmission of 94 per cent at 0-222/i and 67 per cent at 0-186/t, for a sample of crystalline quartz 1 cm. in thickness. Some samples of amorphous quartz have been found to be more opaque than crystalline quartz ; but this may be the result of contamination in melting.

The infra-red transmission of quartz haa been determined by various observers. A characteristic absorption band occurs at about 2-Q5/J.. A sample 1 cm. in thickness is practically opaque (38) to radiations of wavelength greater than 4/i (see Fig. 1).

The absorption, reflection, and dispersion constants of quartz are given in a paper by Coblentz (39), who determined the transmission of samples 3 cm. in thickness. The paper gives also factors for eliminating the absorption in a quartz prism.

In the short wave-lengths the refractive indices of quartz have been determined by